workflow-acquisition-archive

Pass

Audited by Gen Agent Trust Hub on Jun 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The workflow orchestrates the execution of multiple command-line utilities for media processing, including FFmpeg, ExifTool, and GNU Parallel (via the media-batch tool).
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the ingestion of content from various external sources, including web platforms via yt-dlp and network-based video streams (RTSP/NDI).
  • [DATA_EXFILTRATION]: The workflow involves transferring processed media assets to external storage providers such as AWS Glacier, Backblaze B2, and Archive.org.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing untrusted data from external sources.
  • Ingestion points: Media metadata and filenames from web sources (via yt-dlp) and network streams (RTSP/NDI).
  • Boundary markers: None identified for untrusted content processing.
  • Capability inventory: Filesystem writes, command execution via GNU Parallel, and network uploads via cloud tools.
  • Sanitization: Includes a specific recommendation to use yt-dlp output templates (-o "%(id)s.%(ext)s") to prevent malicious characters in filenames from affecting the filesystem.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 27, 2026, 09:54 AM
Security Audit — agent-trust-hub — workflow-acquisition-archive